Hello Jean,
this is something we also find in other notation programs, because the voice
does not only give the stem direction, but also the offset direction for
collisions. The idea is basically: While you go from top/bottom to the center
voices they offset into different directions.
So from a formatting perspective this order makes sense. And the big problem of
a "natural" order is that the behaviour depends ob what other voices there are,
so if you have v1 v2 then v2 would mean lowest voice, in v1 v2 v3 v4 this would
mean the second highest voice.
As you probably can imagine this would be hard to implement and would very
easily lead to unwantes behaviour.
Cheers, Valentin
30.09.2021 10:41:48 Jean Abou Samra <j...@abou-samra.fr>:
> Le 30/09/2021 à 10:04, Valentin Petzel a écrit :
>> Hello Silvain,
>>
>> Note that I specified \voice number, so number is not part of the name, but
>> an
>> argument. But Lilypond does (similar to TeX) interpret \command1 as \command
>> 1
>> (which is the reason why numbers cannot be part of variable names, also
>> similar to how \commandA\commandB do not need to be separated, as well as
>> \command{ ... } is possible).
>>
>> This is the exact reason why I am suggesting this option: For instead of
>> having a fixed number of commands \voiceOne, \voiceTwo, ... you simply have
>> to
>> say \voice1 (\voice 1), \voice2 and you can thus easily use an arbitrary
>> number of voices.
>>
>> Cheers,
>> Valentin
>
> There is also the to me annoying concept
> that \voiceTwo is usually the third voice.
>
> Perhaps we could make \voice n format the
> n-th voice in the normal order?
>
> Jean
